As an open source platform, Kubernetes has become a core component in the digital transformation of enterprises worldwide. Kubernetes minimizes outages and disruptions through self-healing, intelligent scheduling, horizontal scaling, and load balancing. Developers can easily roll out and roll back application versions, whether collaborating in development and test environments or deploying to production. IBM is a key player in Kubernetes, Istio, and Knative open source projects, and IBM developers are also contributing to several high-profile Special Interest Groups.
If you are attending KubeCon Europe 2019 May 20-23 in Barcelona, Spain, check out the following sections to learn how developers at IBM are participating. And even if you can’t attend this time, you can learn about the areas we are focusing on during the changing times in the cloud-native landscape.
The containerd project
A relatively new project with the Cloud Native Computing Foundation (CNCF), containerd is a core component of many cloud platforms and other types of products based on container runtime environments. Docker-compatible container support and virtual machine support in Kubernetes integrations swelled tremendously over the past year from IBM and other cloud platform providers. Mike Brown and Phil Estes from IBM are two of the owners and maintainers on containerd. Mike focuses on the container runtime integration with Kubernetes, and Phil focuses on the core components of containerd. Members of Open Containers Initiative (OCI), Mike and Phil are frequently contribute and maintain the projects. OCI is an important standards-based group that IBM invests in to ensure customer success with hybrid cloud and multi-cloud solutions.
Mike is joined by Wei Fu of Alibaba (recently added as a containerd maintainer), for a session on how users can enhance containerd without needing to first modify containerd’s internals. They cover building custom snapshotters for special storage needs and integrating with custom runtime environments for stronger isolation. Mike highlights the architecture and data flow, which are key to extending containerd’s built in functionality.
The session also explains the internals of containerd, covering its components and dataflows and how external plug-ins work. The presentation discusses using containerd’s smart client API and plug-ins to make new and custom integrations. Mike and Wei demonstrate how various sandbox technologies can be integrated with containerd to work with Kubernetes, including Amazon’s Firecracker and Google’s gVisor. Attendees can expect to leave the talk understanding how they can extend and modify containerd to support enhanced integration for custom production deployments.
Knative is a Kubernetes-based platform to build, deploy, and manage modern serverless workloads. IBM is contributing to the Knative project on several fronts. Greg Haynes is one of the maintainers of the Serving component for Knative.
One of the main advantages of serverless platforms is scale-to-zero functionality, which means application developers have an almost-no cost when the applications are idle. But with this approach, developers face the dreaded cold-start problem: When your application is not in use for an extended period, an activating request can take significantly longer to complete.
Greg, leading the effort to reduce cold-start time in Knative, presents a talk on his cold-start performance work. He discusses low-level performance issues that Knative has encountered with Kubernetes as it tries to deliver sub-second, cold-start time.
Istio is an open platform to connect, manage and secure network of microservices. Istio was jointly launched by IBM, Google, and Lyft. Today IBM is actively collaborating with open community and contributing to Istio open source project.
IBM team is hosting Kubernetes and Istio workshop Get the Cert: Build Your Next App With Kubernetes + Istio hosted by IBM. Join us to learn how Kubernetes and Istio make it easy to bind your app to advanced services like Watson, Blockchain, and IoT. Our developers walk through each step in this hands-on lab, and you can walk away with a certification badge.
The Kubernetes Service Catalog
IBM contributes to and uses the Kubernetes Service Catalog in IBM product offerings. Jonathan Berkhahn, a co-chair of the Service Catalog Special Interest Group (SIG), presents both an introduction and deep-dive sessions about Service Catalog. The intro session covers a basic overview of Service Catalog, what the Open Service Broker (OSB) API is, and why Kubernetes would want to use it. The deep-dive session includes a more in-depth overview of how Service Catalog actually works, some design challenges faced while implementing it, and recent work such as
namespaced resources and future plans for the project.
Jonathan also is participating in the contributor summit and meeting with other members of the SIG to discuss the group’s activities. The SIG plans for the next major release to include transitioning Service Catalog to custom resource definitions (CRDs), implementing support for user-provided services, and updating Service Catalog to the include the most recent features of the OSB API.
The Kubernetes Conformance program
IBM is actively participating in many aspects of the Kubernetes Conformance program. Srinivas Brahmaroutu is leading and driving conformance work with the Cloud Native Computing Foundation community to add more conformance coverage.
Srinivas presents a new proposal on validation suites at the Conformance Deep Dive. In the working group meeting Srini highlights the current progress on the work and also discusses other topics about documenting conformance work for each of the Kubernetes releases.
The IBM Cloud Special Interest Group
The IBM Cloud Special Interest Group (SIG) for Kubernetes focuses on building, deploying, maintaining, supporting, and using Kubernetes on IBM public and private clouds. This group plans many activities in Barcelona.
At the Contributor Summit, Sahdev Zala leads a face-to-face session for current contributors to discuss SIG activities and a future roadmap for IBM Cloud SIG. Sahdev also represents the group at the meet-and-greet session for the new Kubernetes contributors who want to learn and contribute to the IBM Cloud SIG.
At KubeCon, Sahdev joins Khalid Ahmed, Nimesh Bhatia, and Brad Topol for an introduction and deep-dive session about the IBM Cloud SIG. Join us to learn about SIG activities and sub-projects, including the newly created IBM Cloud Provider for Cluster API and ongoing work on the IBM Cloud Provider interface.
The Twelve-Factor App Methodology is a software methodology for building scalable microservice applications that includes best practices for building applications deployed to the web with portability, resilience, and scalability. IBM’s Brad Topol and Michael Elder give an overview of the methodology and describe how developers can leverage the core constructs provided by Kubernetes to support the 12 factors for scalable web apps. This talk includes live demonstrations of how Kubernetes can support Twelve-Factor App Methodology for both newer cloud native applications and legacy enterprise middleware applications that include stateful and transactional workloads.
“Kubernetes in the Enterprise”
The IBM booth at KubeCon hosts a series of talks at the mini-theater. Dr. Brad Topol presents Deploying Kubernetes in the Enterprise on Wednesday, May 22 at 2:30. This session shows how to use Kubernetes to deliver existing applications or more resilient cloud-native applications with speed and efficiency. A book signing event for the O’Reilly book “Kubernetes in the Enterprise” immediately follows the session.
Catch IBMers at the following talks
Monday, May 20
- Get the Cert: Build Your Next App With Kubernetes + Istio, hosted by IBM (additional registration required)
- Lightning Talk: Back to the Future with eBPF – Beatriz Martínez Rubio, IBM
- Lightning Talk: Fake it Until You Make it: Unit Tests with Go-Client Fake Client – Fernando Diaz, IBM
- IBM Cloud SIG Face to Face (F2F) meeting with Sahdev P. Zala
Tuesday, May 21
- Intro: Service Catalog SIG – Jonathan Berkhahn, IBM
- Kubernetes + Encrypted Memory = Security * Privacy – Harshal Patil and Pradipta Banerjee, IBM
- Tutorial: Building Security into Kubernetes Deployment Pipelines – Michael Hough, IBM and Sam Irvine, ControlPlane
- Istio Multi-Cluster Service Mesh Patterns Explained – Daniel Berg and Ram Vennam, IBM
- Learn how to Leverage Kubernetes to Support 12 Factor for Enterprise Apps – Brad Topol and Michael Elder, IBM
- Panel Discussion: Ask Us Anything: Microservices and Service Mesh – Lin Sun, IBM; Jason McGee, IBM; William Morgan, Buoyant; Zack Butcher, Tetrate; and Louis Ryan, Google
Wednesday, May 22
- Sponsored Keynote: What I Learned Running 10,000+ Kubernetes Clusters – Jason McGee, IBM Fellow, IBM
- Building Cross-Cloud ML Pipelines with Kubeflow with Spark & Tensorflow – Holden Karau, Google and Trevor Grant, IBM
- Deep Dive: Service Catalog SIG – Jonathan Berkhahn, IBM
- Strategies to “Kubernetify” Legacy Applications – Sai Vennam, IBM
- Panel Discussion: Leveraging Cloud Native Technology to Transform Your Enterprise – Anni Lai, Huawei; Brad Topol, IBM; Ken Owens, Mastercard; Jeff Brewer, Intuit
- Birds of a Feather: Do You Really Need on Premises Serverless ? – Igor Khapov, IBM
Thursday, May 23
- Intro + Deep Dive: IBMCloud SIG – Sahdev P. Zala, Khalid Ahmed, Brad Topol and Nimesh Bhatia, IBM
- Let’s Try Every CRI Runtime Available for Kubernetes. No, Really! – Phil Estes, IBM
- Intro + Deep Dive: Serverless WG – Scott Nichols, Google and Doug Davis, IBM
- Where’s My Pod, Already? A Tale of Cold-Start Struggles on Kubernetes – Gregory Haynes, IBM
- Istio on Knative Lessons Learned: How Istio is Fit for Serverless Platform – Ying Chun Guo and Iris Ding, IBM
- Intro + Deep Dive: containerd – Wei Fu, Alibaba and Mike Brown, IBM
- Intro + DeepDive: Kubernetes Conformance WG